Levy, et al v. Lexington Cty SC Sch, et al, No. 3:2003cv03093 - Document 270 (D.S.C. 2012)
Court Description: ORDER and OPINION denying 221 Motion for Summary Judgment; granting 222 Motion for Summary Judgment Signed by Chief Judge Margaret B Seymour on 4/12/2012.(mdea )
Download PDF
<![CDATA[
Levy, et al v. Lexington Cty SC Sch, et al Doc. 270 IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF SOUTH CAROLINA COLUMBIA DIVISION R.O. Levy and Betty A. Ethredge, ) ) C/A No. 3:03-3093-MBS Plaintiffs, ) ) vs. ) ) OPINION AND ORDER Lexington County, South Carolina, ) School District Three Board of Trustees; ) and Elton Wilson, in his official capacity ) as Chair of the Lexington County ) Registration and Election Commission, ) ) Defendants. ) ____________________________________) Plaintiffs R.O. Levy and Betty A. Ethredge are black registered voters who reside in the Batesburg-Leesville area of Lexington County, South Carolina, and within School District Three. Plaintiffs bring this action under Section 2 of the Voting Rights Act, 42 U.S.C. § 1973, which provides: (a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in section 1973b(f)(2) of this title, as provided in subsection (b) of this section. (b) A violation of subsection (a) of this section is established if, based on the totality of the circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) of this section in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion to the population. Dockets.Justia.com School District Three is governed by a Board of Trustees comprised of seven members elected at large. Each member sits for a period of four years, and the terms are staggered. Prior to 1994, elections were held annually in the last week of February, and a number of black Trustees sat on the Board. In 1994 the date for holding elections was changed by the South Carolina Legislature. The elections now are held at the time of the November general election in even numbered years. The change was approved by the Justice Department as non-retrogressive.1 Black persons constitute approximately twenty-five percent of the voting age population of School District Three. Plaintiffs contend that the current at-large method of voting during the general election has resulted in dissolution of the minority vote by the majority population since 1994, when elections for the School Board were changed from annually during the last week of February, to November of even-numbered years to coincide with the general election date. The parties appeared before the court for a bench trial on December 5, 2005. At the request of the parties, the trial recessed after two days of testimony. The trial reconvened on March 8, 2006, and concluded on March 10, 2006. On February 19, 2009, the court issued findings of fact and conclusions of law in which the court determined that, based upon analysis of the 1994-2004 school board elections, the applicable Gingles factors2 had been met and that, under the totality of the Section 5 of the Voting Rights Act of 1965 (42 U.S.C. § 1973c) was intended to insure that the gains achieved in minority political participation would not be destroyed through new discriminatory procedures and techniques. Beer v. United States, 425 U.S. 130, 141 (1976) (quoting Senate Report No. 94-295, 19, U.S. Code Cong. & Admin. News 1975, p. 785). “In other words, the purpose of § 5 has always been to insure that no voting-procedure changes would be made that would lead to a retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise.” Id. 1 See Thornburg v. Gingles, 478 U.S. 30 (1985). In Gingles, the Court held that, to establish a violation of the “results” standard of Section 2, a plaintiff is required to show that: (1) the minority group is sufficiently large and geographically compact to constitute a majority in one or more single 2 2 circumstances, black voters in School District Three have been denied equal opportunity to elect School Board members of their choice. The court further concluded that at-large elections for the School Board of School District Three diluted minority voting strength in violation of Section 2 of the Voting Rights Act. The court granted Defendants’ motion for interlocutory appeal on March 27, 2009. On April 17, 2009, the court rejected Defendants’ motion to supplement the record with the results of the 2008 school board election, stating that no party had moved for reconsideration of the February 19, 2009 order, and that the 2008 school board election would be of little benefit to the court absent inclusion of the 2006 election. The court determined that the interests of justice would best be served by closing the record. By opinion dated December 21, 2009, the Court of Appeals for the Fourth Circuit remanded the action, finding that the court’s denial of Defendants’ motion to supplement the recorded “worked an injustice in the particular circumstances.” Levy v. Lexingon County, 589 F.3d. 708, 715 (4th Cir. 2009) (internal punctuation omitted). The Fourth Circuit vacated the court’s order and remanded for further proceedings in which the parties would have the opportunity to present the 2006 and 2008 election results. Id. The Fourth Circuit also took the opportunity to provide guidance to the court with respect to the proper evaluation of the Gingles factors. First, the Fourth Circuit reviewed the court’s evaluation of the third Gingles factor, whether the majority votes as a bloc to enable it to usually defeat the minority’s preferred candidate. The Fourth Circuit observed that the court had “rigidly” member districts; (2) the minority is politically cohesive, i.e., tends to vote as a bloc; and (3) the majority votes sufficiently as a bloc to enable it usually to defeat the minority’s preferred candidate. 478 U.S. at 50-51. The court previously has determined that Plaintiffs have satisfied the first Gingles factor. 3 used a fifteen percentage point differential to determine whether a candidate received “significantly less” minority support than the unsuccessful top minority vote-getter. The Fourth Circuit noted that a fifteen percentage point differential may be appropriate, but that the court should consider the number of candidates and number of seats to be filled in order to determine whether the unsuccessful top minority vote getter received a “significantly higher” percentage of the minority community’s support than did the other candidates. “Only then,” according to the Fourth Circuit, “should the court make a determination of whether the unsuccessful top minority vote-getter received a significantly higher percentage of the minority community’s support than did other candidates.” Id. at 717. Second, the Fourth Circuit held that the court erred in assuming that a person without support from over fifty percent of minority voters cannot be deemed a minority-preferred candidate. The Fourth Circuit noted that in Lewis v. Alamance County, 99 F.3d 600, 614 (4th Cir. 1996), it had considered the treatment of second- and third-place finishers behind a successful, minority-preferred candidate of choice who had received more than fifty percent of the minority vote. The Fourth Circuit observed that in Lewis it had found, in a multi-seat election, [c]andidates who receive less than 50% of the minority vote, but who would have been elected had the election been held only among black voters, are presumed also to be minority-preferred candidates, although an individualized assessment should be made in order to confirm that such a candidate may appropriately be so considered. Levy, 589 F.3d at 717. The Fourth Circuit concluded that, “[i]f such candidates can be considered minority-preferred candidates of choice in an election in which the top vote-getter received more than 50 percent of the vote, we see no reason why a similarly popular candidate–one that would have been elected had the election been held only among African-American voters–should not also be 4 considered a minority-preferred candidate of choice in an election which no candidate received 50 percent or more of the minority vote.” Id. at 717-18. Thus, the Fourth Circuit determined that “in elections in which no candidate receives a majority of the minority vote, a candidate may still be labeled a minority-preferred candidate of choice if that candidate would have been elected had the election been held only among minority voters, so long as an individualized assessment of that candidate supports that conclusion.” Id. at 718. The Fourth Circuit cautioned that the court should ensure that the candidate can be fairly considered to be a representative of the minority community, and that, in addition to the “bare statistics,” the court may consider testimony from observers and the candidates themselves to determine whether those candidates may be labeled minority-preferred candidates of choice. Id. at 718, n.14. Third, the Fourth Circuit reproved the court for failing to sufficiently explain why it favored the statistical analysis of John C. Ruoff, Plaintiffs’ expert, over that of David Epstein, Ph.D., Defendants’ expert. The Fourth Circuit noted “problems” with the data provided by Dr. Ruoff that “did not appear to have been considered” by this court. The Fourth Circuit observed that Dr. Ruoff’s data was “suspect” and that this court had failed to acknowledge or explain what appeared, on review, to be the limitations of Dr. Ruoff’s report. Id. at 718-19. The Fourth Circuit next addressed the second Gingles factor, whether a minority group is politically cohesive. The Fourth Circuit stated that this court erred in considering support only for those candidates it already had classified as minority-preferred candidates of choice, i.e., those with over fifty percent minority support. The Fourth Circuit found that this court erred in assuming that a minority-preferred candidate of choice needed to have over fifty percent of minority support, because such an approach essentially “guaranteed a finding of cohesiveness.” Levy, 589 F.3d at 720. 5 The Fourth Circuit observed that the 2002 electoral results, wherein no candidate achieved more than fifty percent of the minority vote, should not have been disregarded, because “the 2002 results presumably demonstrated a lack of political cohesion–the very factor the district court sought to establish under the second Gingles factor.” Id. at n.18. The Fourth Circuit further elucidated for this court the distinction between political cohesion and racial polarization. Levy, 589 F.3d at 720. The Fourth Circuit noted that minority voters who are racially polarized may still lack political cohesion, and instructed the court to reconsider this issue on remand. Subsequent to a status conference on February 9, 2010, the parties submitted additional briefing and supplemental expert reports addressing the concerns enumerated in the Fourth Circuit’s opinion remanding this case. The parties filed cross-motions for summary judgment on May 17, 2010. Plaintiffs and Defendants each filed a response in opposition to the opposing party’s motion on May 28, 2010. Plaintiffs and Defendants each filed a reply brief on June 4, 2010. The court held a hearing on June 14, 2010. At that time, the court indicated that it wanted to hear testimony of the expert witnesses, Dr. Ruoff for Plaintiffs and Dr. Epstein for Defendants. An evidentiary hearing was held on July 20, 2010 and July 30, 2010, at which time both experts testified as to their findings and conclusions. After reviewing the new information from the evidentiary hearing, the court was aware that both Plaintiffs’ and Defendants’ experts have tended to change definitions and methodologies used throughout the case. In addition, Defendants strongly objected to Dr. Ruoff’s use of a form of robust regression known as “MM estimator” to address outliers, specifically the Ridge Road precinct.3 Dr. Ridge Road is a particularly large, overwhelmingly white precinct, that is an outlier and thus required special treatment in the regression analysis. At the time of the 2004 election, Ridge Road contained 1,087 voters, only 22 (2%) of whom are black. Plaintiffs’ expert, Dr. Ruoff, contends that 3 6 Ruoff also created a “5% difference rule” which he applied to determine whether to utilize the robust regression method over ecological regression analysis in establishing the level of minority support for a candidate. See, e.g., ECF No. 238, 16. The experts’ differing treatment of the Ridge Road precinct as an outlier provided the court with little guidance, and it appeared that the determination of how to treat the Ridge Road precinct could be critical in making a determination as to minority support for various school board candidates. The court recognizes that the statistical evidence presented by both parties is problematic not just because of the varying methodologies employed by the experts and the various interpretations of the results thereof. The nature of the School District Three at-large elections, where numerous candidates vie for only three or four seats, and where voters may cast all, none, or any number of their available votes during an election, presents unique challenges in analyzing the data. Accordingly, the court engaged the services of a statistical expert to educate the court regarding the methodologies utilized by both Dr. Ruoff and Dr. Epstein.4 See 5 U.S.C. § 3109 and Guide to Ridge Road’s presence distorts the regression analysis and overestimates minority support for white candidates. Thus, he removed Ridge Road from the analysis. Defendants’ expert, Dr. Epstein, contends that under proper assessment, outliers would not be removed from the analysis at all. Specifically, Dr. Epstein opined that the Ridge Road observations are not invalid; they simply reflect less racially polarized behavior among white voters. The court acknowledges the invaluable instruction provided by Kevin M. Quinn, Ph.D. Dr. Quinn currently is Professor of Law at the University of California, Berkeley School of Law. He received both an A.M. and Ph.D. in political science from Washington University in St. Louis and was an Associate Professor of Government at Harvard University before joining the faculty at Berkeley in 2009. Dr. Quinn has written extensively on judicial decision making and statistical methodology, and is a three time winner of the Gosnell Prize for excellence in political methodology. He also serves as Associate Editor for the Journal of the American Statistical Association. See http://www.law.berkeley.edu/kevinmquinn.htm (April 11, 2012). Dr. Quinn also is a Principal with Principia Empirica, LLC. Dr. Quinn was retained solely to educate the court. Dr. Quinn did not make findings or render an opinion as to the merits of the expert reports or testimony submitted by the parties. See ECF No. 259. 4 7 Judiciary Policy, Vol. 14, Ch. 5, § 520, Expert and Consultant Services. On September 9, 2011, the court requested the parties and their respective experts to answer the following questions: 1. All of the estimation methods used by the experts assume the absence of what is sometimes called “aggregation bias.” The lack of aggregation bias implies: • the precinct-specific fraction of blacks who support a particular candidate in a particular election is not systematically related to the precinct-specific fraction of voters who are white (or black) in that election. • the precinct-specific fraction of whites who support a particular candidate in a particular election is not systematically related to the precinct-specific fraction of voters who are white (or black) in that election. Are there reasons to believe that either (or both) of these conditions are not satisfied for some race, candidate, precinct, and/or election combinations? If there are reasons to believe that these conditions are not satisfied for all precincts in a particular election, are there reasons to believe that these conditions are satisfied, at least approximately, for a subset of the precincts in the election in question? If so, what are these precincts and what can be inferred about black and white voting behavior in these “no aggregation bias” precincts? More specifically, is it appropriate to use ecological regression for the subset of precincts that satisfy the assumptions underlying ecological regression in a particular election? Could the resulting estimates be combined with the observed aggregate data from the precinct(s) left out of the regression analysis to make inferences about black and white support for candidates in the district as a whole? 2. What is the precise scientific justification for performing ecological regression using weighted least squares with weights based on the size of the voting population in each precincts? Please justify the use of weighted least squares in light of criticisms of such an approach raised by King (1997). If weighted least squares is not justified in the context of this case, new estimates of the relevant quantities of interest (black and white support for the various School District Three candidates) should be computed using regression models, such as ordinary least squares regression, MM-regression, etc., that do not differentially weight observations based on the size of the voting population. ECF No. 260. The parties filed responses to the court’s questions on October 7, 2011. Dr. Ruoff stated that 8 the vast majority of contests for white candidates in School District Three demonstrate substantial variance, which strongly suggested an aggregation bias problem. Dr. Ruoff noted that the principal problematic precinct is Ridge Road, and that “[i]n those contests in which Ridge Road creates the aggregation bias in the estimates from all precincts, ecological regression for the subset of no aggregation bias precincts provides appropriate and reasonable estimates for those precincts[,]” but was problematic in other scenarios. ECF No. 264-1, 3. Dr. Ruoff created new estimates for the 1994-2008 elections using an average of MM estimates and ecological inference estimates. ECF No. 264. Dr. Epstein responded that the ecological inference technique is preferable because it minimizes the effect of aggregation bias on statistical estimates. ECF No. 265-1. On October 25, 2011, Defendants filed objections to Plaintiffs’ reply, contending that Dr. Ruoff improperly had developed an entirely new statistical analysis that was outside the scope of the questions propounded by the court. Plaintiffs filed a response in opposition to Defendants’ objections on November 1, 2011, to which Defendants filed a reply on November 15, 2011. DISCUSSION The electoral contests at issue in this case are multi-candidate, at-large elections without slotted seats. In these contests, each voter can cast multiple votes but need not cast all of the votes allocated to him or her. As discussed hereinabove, none of the models / estimation methods used by Dr. Ruoff or Dr. Epstein is designed specifically for such contests. The court declines to utilize the “5% difference rule” suggested by Dr. Ruoff. The “5% difference rule” is without scientific basis and is not considered further. See ECF No. 246, 5 (wherein Dr. Ruoff admits that the 5% difference rule is an independent creation and not part of robust regression); ECF No. 246, 81-82 (wherein Dr. Epstein testifies that he has never seen a “5% 9 difference rule” in the literature or in practice). The statistical models/estimation methods used by Dr. Ruoff (unweighted ecological regression, weighted ecological regression, ecological regression using MM estimation) are not without problems. The estimates produced by such methods can, in some circumstances, be logically impossible. For example, such methods can produce results that, if taken literally, would imply that over one hundred percent of African-American voters supported a particular candidate. See ECF No. 246, 67-68 (wherein Dr. Epstein notes that under traditional ecological regression, numbers at face value can reflect over one hundred percent support rates or less than zero percent support rates); ECF No. 265-1, 5 (wherein Dr. Epstein notes criticisms of ecological regression with weighting based on precinct turnout because such method can produce estimates of the percent of voters casting ballots for a given candidate as either under zero percent or over one hundred percent). The statistical models used by Dr. Epstein (King’s “EI” method and the ecological inference program developed by Kosuke Imai, Ying Lu, and Aaron Strauss) also are problematic. These models are designed for electoral contests with two candidates in which each voter has a single vote. King has developed extensions of his model that could be appropriate for multicandidate contests in which each voter can cast only a single vote. However, even when applied to single-vote contests, these methods are not without potential drawbacks. Similar issues exist for the ecological inference method of Imai, Lu, and Strauss, that is, the School District Three contests at issue are not single-vote contests or their functional equivalent. Consequently, the methods used by Dr. Epstein are not optimal for analyzing these contests. A small number of academic articles have explicitly addressed the analysis of multicandidate 10 at-large elections without slotted seats. One of the most recent is by D. James Greiner,5 “ReSolidifying Racial Bloc Voting: Empirics and Legal Doctrine in the Melting Pot,” 86A Indiana Law Journal, 447-97 (2011). The court finds the following persuasive: A particular challenge is whether any method can tackle contests in which each voter can cast more than one vote for a group of candidates, that is, contests in an at-large election system without slotted seats such as the one used for the Boston City Council. The problem is more difficult than it might appear at first blush because there is an additional, hard-to-model aspect of voting behavior that we do not observe, namely, the number of votes each voter casts. In my view, . . . it may be best to retreat to simplicity in this situation by using correlation coefficients and other blunt measures. Id. at 470, n.117. Correlation coefficients are deterministically related to ecological regression coefficients.6 Because the experts in this case have not provided correlation coefficients for all candidates in all elections, the court will make use of the ecological regression estimates of Dr. Ruoff. A final issue concerns whether the ecological regression results should be from unweighted ecological regressions or weighted ecological regressions, where the weights are based on the size of the voting population. As has been made clear by King,7 weighted ecological regression is a method for dealing with unequal variances of the error terms in the regression (heteroscedasticity). D. James Greiner is an Assistant Professor of Law at Harvard Law School. He received a J.D. from the University of Michigan in 1995 and a Ph.D. in statistics at Harvard University in 2007. His research focuses on statistics and litigation, including ecological inference models used in Section 2 cases. See http://www.law.harvard.edu/faculty/directory/index.html?id=705 (April 10, 2012). 5 Consider a simple linear regression yi = a + bxi + ui. The estimated slope coefficient b is equal to the sample correlation between x and y times the sample standard deviation of y divided by the sample standard deviation of x. David A. Freeman, Statistical Models: Theory and Practice 19-20 (New York: Cambridge University Press 2005). 6 Gary King, “A Solution to the Ecological Inference Problem: Reconstructing Individual Behavior from Aggregate Data” 66-65 (Princeton Univ. Press 1997). 7 11 This is distinct from the related notion of survey weighting in which weighted averages are used to produce more accurate estimates of population quantities from unequal probability samples. As demonstrated by Dr. Ruoff in his report of October 7, 2011, there is no evidence to suggest that the error variances vary with the size of the voting population. See ECF No. 264-1. The court will make use of the unweighted ecological regression results presented by Dr. Ruoff. These can be found in Response of John C. Ruoff, Ph.D. dated October 7, 2011, Attachment JCR-2 (Racial Bloc Voting Estimates) under the column heading of “Unweighted ER” and subheading “% Turnout.” ECF No. 264-1, 27-28. One determination that requires judicial discretion regarding candidates of choice is what constitutes a “significantly higher” gap between the top minority vote-getter and another candidate. This notion of “significance” is distinct from any sense of “statistical significance.” There is no unique, principled method for determining whether one voteshare is “significantly higher” than another. As noted hereinabove, the Fourth Circuit has cautioned this court that this determination should depend on the number of candidates and the number of seats to be filled. In addition, an important consideration is the extent to which black voters use fewer than their allocated number of votes. As explained in Collins v. City of Norfolk, 883 F.2d 1232, 1239 (4th Cir. 1989): In 1974, 21 candidates sought the 4 open seats on the council. In 1980, 12 candidates vied for 3 open seats. In 1974, each voter–black and white–could cast four votes and in 1980 three votes. If black voters exercised their right to cast all of their allotted votes, they ran the risk that their second and third choices would be declared their preferred candidates. Only by single-shot voting–withholding all votes save for their first choice, and forfeiting the opportunity to cast all votes allotted to each voter could the minority be assured that its second and third choices would not be declared its preferred candidates. The secret ballot prevents the court from knowing how many of the unused ballots are the 12 result of black voters using, for example, two of their allocated four votes, and how many of the unused ballots are the result of black voters using zero or one of their allocated four votes. Further, the criteria laid out by the Fourth Circuit for determining candidate of choice status depend on white voting patterns even when black voting patterns are held fixed. Consider a situation where there are 100 black voters and 200 white voters. Seven candidates (A, B, C, D, E, F, G) are competing for three open seats, and each voter can vote for up to three candidates. Assume that the court can determine the number of black voters who voted for each candidate and the number of white voters who voted for each candidate. Tables 1 and 2 depict vote counts under two hypothetical scenarios. Candidate Black Votes White Votes Total Votes Elected Black Cand. of Choice A B C D E F G 97 52 51 40 25 20 15 180 120 110 60 60 50 20 277 172 161 100 85 70 35 Yes Yes Yes No No No No Automatic Yes Automatic Yes Automatic Yes No No No No Table 1: Scenario 1 Candidate Black Votes White Votes Total Votes Elected A B C D E F G 97 52 51 40 25 20 15 20 50 60 60 110 120 180 117 102 111 100 135 140 195 Table 2 Scenario 2 13 No No No No Yes Yes Yes Black Cand. of Choice Automatic Yes Yes if 0.97 not “signif. higher" than 0.52 Yes if 0.97 not “signif. higher" than 0.51 No No No No The black vote counts are exactly the same in the two scenarios. The only thing that changes is white voting behavior. In Scenario 1 white voters tend to vote for the same candidates as black voters. As a result, the candidates with majority black support all win seats. Consequently, Candidates A, B, and C are automatically classified as black candidates of choice. In Scenario 2, white support is highest for candidates least preferred by blacks. As a result, none of the candidates with majority black support wins a seat. Consequently, only Candidate A, the top minority vote getter, is an automatic black candidate of choice. Candidates B and C could be black candidates of choice if the ninety-seven percent support that Candidate A received from black voters is deemed to be not “significantly higher” than the fifty-two percent and fifty-one percent support that Candidates B and C received. Dr. Epstein’s interpretation of the Fourth Circuit’s standard for determining whether a candidate is a black candidate of choice depends on whether the candidate won or lost the election. If the court were to apply Dr. Epstein’s construction of Levy, the result could be that a candidate would be deemed a candidate of choice for black voters because the candidate’s level of white support increases. The court disagrees with Dr. Epstein’s interpretation of Levy. As suggested by Dr. Greiner, the court will “retreat to simplicity” and utilize the unweighted ecological regression figures prepared by Dr. Ruoff to determine whether minority voters in School District Three have been able to elect candidates of choice to the school board, or whether the majority votes as a bloc to enable it to usually defeat the majority’s preferred candidate. Law/Analysis The court first will turn to the third Gingles factor, whether the majority votes as a bloc to enable it to usually defeat the minority’s preferred candidate. For the reasons set out hereinbelow, 14 the court concludes that the majority does not vote as a bloc to defeat the minority candidates of choice in School District Three. The parties generally agree that three minority candidates of choice have been elected since 1994: Asbill in 1994 (white; 82% of the black vote); Lester in 2004 (black; 94% of the black vote); and Drafts in 2008 (black, 71 percent of the black vote). Defendants contend, and Plaintiffs dispute, that the following candidates also were minority candidates of choice between 1994 and 2008: Berry in 2002; Kennedy and Fox in 2004; Berry, Burgess, and Rikard in 2006; and Kennedy, Fox, and Caughman in 2008. The court will focus on the 2002, 2004, 2006, and 2008 elections using unweighted ecological regression estimates to determine whether evidence supports Defendants’ assertion that these individuals constitute minority candidates of choice. In 2002, six candidates vied for three seats on the School District Three school board. Utilizing Dr. Ruoff’s unweighted regression analysis for the 2002 election, voters apportioned their votes for as follows: 2002 Black White Result Padgett 47.5 29.9 (White, Lost) Berry 47.4 39.4 (White, Won) Etheredge 42.9 42.6 (Black, Lost) Burgess 22.8 49.4 (White, Won) Rikard 22.6 45.8 (White, Won) Hendrix 12.1 34.6 (White, Lost) Berry was a successful candidate. However, neither Berry, Padgett, nor Etheredge received a majority of the minority vote. Under Levy, candidates who would have been elected had the election been held only among black voters are presumed to be minority-preferred candidates, 15 subject to an individualized assessment to confirm that such a candidate may appropriately be considered to be a representative of the minority community. In this case, Berry, Padgett, and Etheredge would have been elected had the election been held only among black voters. Berry won a seat on the school board. The court concurs with Defendants that Berry is presumptively a candidate of choice of the minority voters, subject to an individualized assessment. Padgett and Etheredge also would be considered to be presumptive candidates of choice; however, they lost the election to Burgess and Rikard. In 2004, six candidates vied for four seats on the School District Three school board. Voters apportioned their votes as follows: 2004 Black White Result Lester 86.1 36.0 (Black, Won) Kennedy 46.3 37.3 (White, Won) Fox 46.2 58.2 (White, Won) Padgett 38.9 45.8 (White, Lost) Rowe 37.2 22.8 (Black, Lost) Koon 28.5 61.3 (White, Won) Lester received a majority of the minority vote and won. Kennedy and Fox did not receive a majority of the minority vote. However, under Levy, Kennedy, Fox, and Padgett would have won had the election been held solely among black voters. Padgett was unsuccessful, having been defeated by Koon. Thus, as Defendants assert, Kennedy and Fox are presumptive candidates of choice, subject to an individualized assessment. In 2006, five candidates competed for three seats on the School District Three school board. Voters apportioned their votes as follows: 16 2006 Black White Result Powell 59.5 22.8 (White, Lost) Berry 52.0 46.4 (White, Won) Burgess 47.3 55.1 (White, Won) Price 40.1 41.9 (White, Lost) Rikard 39.1 47.6 (White, Won) Powell received a majority of the minority vote, but was defeated. Berry received a majority of the minority vote and was successful. The court concurs with Defendants that, under Levy, Berry was a candidate of choice of the black community. Burgess did not receive a majority of the minority vote and finished behind an unsuccessful candidate of choice, Powell. Under Levy, the court must determine whether Burgess received “significantly less” minority support than the unsuccessful top minority vote-getter. In this case, Burgess received nearly a majority of the minority vote and was within 12.2% of the unsuccessful top vote-getter. Stated differently, Burgess received approximately 79.5% of the vote received by Powell in the competition for three seats on the school board. The court cannot say that Burgess received “significantly less” minority support than Powell. Therefore, Burgess is a candidate of choice of the minority community. Rikard also did not receive a majority of the minority vote. However, Rikard finished last among the five candidates with only 39.1% of the minority vote; the differential between Rikard and Powell is 20.4%; and Rikard received only 66% of the vote received by Powell in the competition for three seats on the school board. The court concludes that Rikard received “significantly less” support than did Powell. Contrary to Defendants’ assertion, Rikard was not a candidate of choice in the 2006 election. Powell was a candidate of choice of the minority voters, but was not elected. 17 In 2008, seven candidates competed for four seats on the school board. The voters apportioned their votes as follows: 2008 Black White Result Drafts 71.2 28.0 (Black, Won) Kennedy 39.0 40.9 (White, Won) Fox 34.7 34.8 (White Won) Gambrell 22.4 22.3 (White, Lost) Goff 19.9 34.6 (White, Lost) Thomas 8.9 23.2 (Black Lost) Caughman 5.7 53.9 (White, Won) Kennedy and Fox finished behind a successful minority candidate of choice. They did not receive a majority of the minority vote; however, they would have been elected had the election been held solely among black voters. Thus, Kennedy and Fox are presumptive candidates of choice, subject to an individualized assessment. Caughman did not receive a majority of the minority vote and would not have been elected had the election been held solely among black voters. The court concludes that, contrary to Defendants’ assertion, Caughman was not a candidate of choice of the minority community during the 2008 election. Instead, under Levy, Gambrell would have been elected had the election been held solely among black voters. Utilizing the unweighted ecological regression estimates, the court concludes that Berry in 2006 was an automatic candidate of choice of the minority community. Thus, between 1994 and 2008, the minority community elected four clear candidates of choice: Asbill in 1994 Lester in 2004 Berry in 2006 18 Drafts in 2008 The court further concludes that the following candidates are presumptive candidates of choice, subject to an individualized assessment: Berry in 2002 Kennedy and Fox in 2004 Burgess in 2006 Kennedy and Fox in 2008 The question becomes, then, whether Berry, Kennedy, Fox, and Burgess can be fairly considered to be representatives of the minority community. As an initial matter, the court notes that Berry received fifty-two percent of the black vote in 2006 when he ran for re-election to the School District Three school board, and became an automatic candidate of choice. Further, the minority support for Burgess in 2006 was comparable to the minority support for Berry, who won, and not “significantly less” (-12.2%) than the minority support for Powell, who lost. Kennedy and Fox present a more difficult issue. The court has observed hereinabove the difficulties of determining candidates of choice when minority voters do not cast all their votes. The wide disparity between the percentage of black vote in 2004 for Lester (86.1%) and for Kennedy (46.3%) and Fox (46.2%), and between the black vote in 2008 for Drafts (71.2%) and for Kennedy (39.0%) and Fox (34.7%) demonstrates that many black voters may not have utilized all their votes. As articulated in Levy, the court may consider testimony from observers and the candidates themselves to determine whether those candidates may be labeled minority-preferred candidates of choice. Defendants have provided declarations of Berry, Kennedy, and Fox as attachments to their motion for summary judgment. • Berry states that he has lived in the Batesburg-Leesville area his entire life. He states that since 1981 he has lived in a section of town that is racially 19 diverse. His children have attended public schools in School District Three. Berry attests that he has been involved in a number of racially diverse community activities, including girls softball, baseball, and co-ed basketball in the local recreational league. Berry states that he got involved with the school board when the District was facing major budget cuts from the state legislature. Berry avers that he sought support from all voters in School District Three, including black, white, and Hispanic. Berry states that he has supported programs that have shown positive results and benefitted minority students. See generally ECF No. 222-4. • Burgess attests that she has lived in District Three for all but approximately eight years. She lives in a racially diverse neighborhood. Her children attended schools in the Batesburg-Leesville school system. Burgess states that one issue she campaigned on was preparing students for life after high school, whether it be for college, an associate degree, technical college, or vocatioal program. She avers that she sought support from both white and black voters, including discussing her goals with persons through her job, at the grocery store, sporting events, and any community event where she had the opportunity to seek support. See generally ECF No. 222-5. • Kennedy notes that minority students in School District Three are disproportionately affected by academic underperformance. Kennedy notes that during his tenure on the school board, significant effort and financial resources have been devoted to the creation and implementation of numerous academic programs specifically designed to address the needs of underperforming students. Kennedy avers that he is acutely aware of and committed to addressing the concerns of minority students in the school district. Kennedy states that he believes he has received substantial support from minority voters in part because of the school board’s efforts in bringing about and supporting academic programs. He also attests that he and his family have been part of the local community for generations and he has relationships with many members of the minority community. He further states that he received black support for his candidacy to the South Carolina House of Representatives. Kennedy avers that he advocates for programs that help both minority students and the minority community as a whole. See generally ECF No. 222-2. • Fox states that he was born and raised in Batesburg-Leesville. He states that he lives in a racially diverse neighborhood. His children attended public schools in School District Three. Fox attests that he has participated in racially diverse community activities and that he presented a eulogy for an African-American gentleman who had been his neighbor and dear friend for Fox’s entire life. Fox avers he was active in School Improvement Councils 20 and PTAs prior to running for the school board. Fox states that he sought support from both white and black voters in School District Three, and that he is well known in both the white and black communities. Fox further states that he has always supported programs that benefit the minority community. See generally ECF No. 222-3. The court is aware that it discussed the Senate Report factors in its Findings of Fact and Conclusions of Law filed December 19, 2009, and found the existence of a history of official discrimination; lingering effects of discrimination; and a failure of elected officials to respond to needs of the minority group. ECF No. 173, 31-36. The court observed that Fox and Berry testified to views inconsistent with other facts tending to establish the existence of the Senate Report factors. For example, Fox opined that he believed the black community in Lexington County did not suffer in any way from the effects of past discrimination. Berry similarly testified that he believed that blacks do not suffer any present day effects from past discrimination. Notwithstanding these statements, these individuals also testified that they have the ability to respond to the needs of minority students in School District Three. For example, Kennedy testified at the bench trial that his father owned a men’s clothing store that served both black and white customers. Kennedy started working at the store at a young age and worked through high school serving the store’s customers. ECF No. 150, 188-89. Kennedy attended School District Three schools from kindergarten until he graduated from high school. Id. at 190. Kennedy coached his son’s Little League and daughter’s interracial softball teams, and became close to a number of black and white students. Id. at 192. Kennedy has a small town general practice where he accepts black, white, and Hispanic clients. Kennedy estimated that his clients were approximately sixty percent white, forty percent black, and a small percentage Hispanic. Id. at 193. Kennedy considers himself to be a candidate of choice because his family is well known in both the white and black 21 communities; he represents numerous minority clients and tries to do a good job for them, and feels that the clients vote for him; and he has tried to make an active difference on the school board and be responsive and receptive to both black and white persons who call him with concerns. Id. at 20809. Fox testified at the bench trial that his senior year in high school he received an award for having worked to promote racial harmony during the period of school integration in the BatesburgLeesville area. He also served on a pre-consolidation subcommittee prior to the consolidation of the towns of Batesburg and Leesville in 1992. ECF No. 150, 85-86. Fox testified that he lives in an integrated neighborhood. Id. at 89. He also testified that all races are well represented and supportive of the school system, school activities, and athletics. Id. at 90. Fox testified that the school board has implemented a mentor program for character building, a “reading buddies” program to improve reading levels at all grades, and after school programs to help children who were scoring below basic to move into the basic categories. Id. at 92-93. Fox denied ever witnessing the school board act in a way that would provide less opportunity for minority students. Id. at 95. Berry testified that he was in high school during desegregation. He was chosen by his peers to serve on a biracial committee that had the function of trying to resolve problems that arose between the students because of desegregation. ECF No. 150, 137-38. Berry believes that his experience on the committee shaped his reputation with black people in Batesburg-Leesville. Id. at 139. According to Berry, a number of biracial and Hispanic persons attend his church. Id. at 142. He considers school desegregation to have been a good thing. Id. at 144. Based on all the evidence now contained in the record, and utilizing the Levy analysis as instructed by the Fourth Circuit, the court concludes that Berry, Burgess, Kennedy, and Fox 22 constitute representatives of the minority community entitled to candidate of choice status. The court finds, and the parties agree, that minority voters in School District Three were unable to elect candidates of choice in the 1996, 1998, and 2000. However, recent elections are the most probative in determining vote dilution. United States v. Charleston County, 365 F.3d 341, 350 (4th Cir. 2004). The court finds that since 2002, minority voters have elected candidates of choice in each election: Berry in 2002; Lester, Kennedy, and Fox in 2004; Berry and Burgess in 2006; and Drafts, Fox, and Kennedy in 2008. Minority candidates of choice lost two out of three seats in 2002 (Padgett with 44% of the minority vote and Etheredge with 37% of the minority vote); one out of four seats in 2004 (Padgett with 40% of the minority vote); one out of three seats in 2006 (Powell with 54% of the minority vote); and one out of four seats in 2008 (Gambrell with 24% of the minority vote). Of those losses, only Powell in 2006 received a majority of the minority vote but lost to Rikard, who was not a candidate of choice of the minority community. Stated differently, out of the last four elections–2002, 2004, 2006, 2008–fourteen seats on the school board were open, and minority voters elected nine candidates of choice, for a sixty-four percent success rate. Moreover, the tables included hereinabove demonstrate that in 2004 Kennedy received a greater percentage of support from black voters (46.3%) than from white voters (37.3%). Fox received comparable support to Kennedy in 2004 from black voters (46.2%), although he received greater white support (58.2%) than did Kennedy. In 2006, Berry received a greater percentage of support from black voters (52.0%) than white voters (46.4%). In 2008, Kennedy and Fox received nearly equal support from black and white voters (Kennedy: 39.0% black vote / 40.9% white vote; Fox: 34.7% black vote, 34.8% white vote). The court find that the record as a whole does not support a finding that, at least in the 2002, 2004, 2006, and 2008 elections, the majority voted 23 sufficiently as a bloc to enable it usually to defeat the minority’s preferred candidate. The court concludes that Plaintiffs have failed to establish the third prong of the Gingles test. The court finds it unnecessary to discuss the second prong of the Gingles test. Plaintiffs have failed to satisfy their burden of establishing a cognizable violation of Section 2 of the Voting Rights Act. CONCLUSION For the reasons stated, Plaintiffs’ motion for summary judgment (ECF No. 221) is denied. Defendants’ motion for summary judgment (ECF No. 222) is granted, and the case dismissed, with prejudice. IT IS SO ORDERED. /s/ Margaret B. Seymour Chief United States District Judge Columbia, South Carolina April 12, 2012 24
]]>
Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You
should read the full case before relying on it for legal research purposes.
This site is protected by reCAPTCHA and the Google
Privacy Policy and
Terms of Service apply.
